MarketWatch: Smart Speakers Are Still Pretty Dumb — Here’s What Needs To Happen
This article was co-authored with Sachin Maini
It’s been four years since Amazon.com’s Echo first hit the market, and so far smart speakers haven’t amounted to much more than glorified radios that can control the lights.
At the very least, these devices were supposed to have changed the way we shop. But a damning report in The Information recently revealed that of the 50 million owners of an Alexa-enabled device, only 2% have used their voice to make a purchase even once — and of those, only 10% made a second purchase. Study after study shows that people continue to use smart speakers primarily for the basics: music, weather, and the news.
While smart speakers are popular and entertaining, at the end of the day they just aren’t that useful. They’re still awkward to interact with, they can’t handle complexity very well, and features with a lot of potential, like push notifications, are still in a nascent stage. As a result, each of the major technology companies in the market — Amazon, Google, and Apple in the US, and Baidu, Alibaba and Tencent in China — have been competing on either price or sound quality.
That leaves us with a fiercely competitive market full of devices that all pretty much do the same thing: not very much.
However, there are a series of applications in the health and security verticals which could transform these devices from a “nice-to-have” to an absolute “must-have” product. Whichever company is able to deliver credible innovations in these two areas will break through the noise. Doing so will require companies to go beyond merely adapting ideas that have already proven to work on mobile phones, but instead creating entirely new concepts that are native to smart speakers and built with a voice-based interaction in mind.
Take health, for example. Contrast the difference between a text-based calendar alert reminding you to take your medications to an automated conversational check-in that asks how your treatment is going and whether you’re experiencing any issues or unexpected abnormalities. Without any additional effort on your part, your responses could be automatically uploaded to a health record and shared as necessary with your doctors and family members to review. That would amount to far more than an incremental improvement.
This isn’t just theoretical; experiments in hospitals have already shown that voice-based assistants can be effective in helping physicians document their procedures, reducing cognitive load for surgeons during operations, and even in aiding patient care.
Promising telemedicine apps for voice platforms, which can connect you with a medical professional or direct you to nearby clinics with the shortest wait times, already exist. The next step is to go from informing to reacting. Technology companies and software developers have a clear opportunity to make medical alert systems drastically more effective. Apple took strides toward this with its Series 4 watch, which is capable of detecting a fall or a heart abnormality. Smart speakers can reinvent the way we connect to care when we need it, especially in situations like mental health emergencies where timely intervention could be lifesaving.
So far, the industry’s efforts have been underwhelming. Of the thousands of wellness and fitness-related skills, actions, and apps currently available across the major platforms, the vast majority are plagued with quality and accuracy issues. That damages the credibility of the whole ecosystem. With health, the stakes are high, so accuracy is extremely important.
Providing more sophisticated tools is even harder. Adding to that, the industry is heavily regulated, and these new tools would require a different sensitivity to personal data than the tech giants are used to.
These devices can also do a lot more to help us to stay safe, get the right information in a crisis, and respond more effectively to emergencies.
Right now, Google and Amazon are gearing up to battle over the $50 billion home security market. Both companies have launched a suite of cameras, doorbells, and motion sensors with the smart speaker as the hub. The offerings, however, leave a lot to be desired. The competitive focus has been on re-creating what the existing security equipment companies already do, at the cost of developing new approaches based on what’s natural and intuitive to voice.
That’s a missed opportunity because voice-based communication can be very effective in communicating immediate and important information. It makes more sense for a smart speaker to verbally announce when a person is at the front door than to get an alert on your phone. There could even be a security mode activated by a specific wake-word such as “Help” or “Danger” which would apply a different set of privacy practices, triggering automatic recording and streamlined communication with emergency services.
Along the same lines, smart speakers could become invaluable tools for disaster preparedness and response. Local governments and services like NextDoor can’t reach more than a small segment of the population. Just as mobile carriers developed the infrastructure to push Amber alerts and warnings, our Amazon Echos and Google Homes could reach out in the wake of a fire or natural disaster to ask if anyone needs assistance. A voice-based interface may be a better way to reach people than text messages or social media check-ins.
It’s not going to be more podcasts, better music, or increasingly sophisticated games that will get to us a point where smart speakers are in every home. What will get us there is a relentless focus on areas where voice-activated assistants can improve life in a meaningful way.